Extract Information with Azure Content Understanding in Foundry Tools AI-102 Practice Questions (Page 4)

Questions 16–20

1. 16expert · hard

   A media company processes press releases that arrive as text files, PDFs, and audio recordings. They need to extract the event date, the location, and the main topic, and generate a summary. They have a constraint: the pipeline must process all three formats in a single run and the location may be missing in some releases. They also want to avoid false positives when the location is absent. What should they do?

   • ACreate a single custom pipeline that accepts text, PDF, and audio, with entity extraction for date and location (location optional) and summarization.
   • BCreate a single pipeline with OCR and entity extraction, and manually generate summaries.
   • CCreate separate pipelines for each format, each with entity extraction and summarization.
   • DCreate a single pipeline with only entity extraction and summarization, and infer the main topic from the summary.
   Reveal Solution Discussion Select an answer first
2. 17foundation · easy

   In Azure Content Understanding Foundry Tools, what is the first step when processing a new batch of content?

   • AIngest the content into the system
   • BClassify the content
   • CSummarize the content
   • DExtract entities from the content
   Reveal Solution Discussion Select an answer first
4. 18application · medium

   A real estate agency receives property brochures in PDF format. Each brochure contains photos of the property, a description paragraph, and the address. They want to extract the photos, the address, and the description text for their listing database. Using Azure Content Understanding in Foundry Tools, what should they configure?

   • AA document classification pipeline that categorizes the brochure and extracts the address.
   • BA summarization pipeline that generates a summary including the address and photo descriptions.
   • CA custom extraction pipeline with image extraction, entity extraction for the address, and OCR for the description.
   • DA table extraction pipeline to pull the address and description from any tables in the brochure.
   Reveal Solution Discussion Select an answer first
5. 19application · medium

   A utility company receives scanned meter reading sheets that include handwritten numbers, a small table of previous readings, and a logo image. They need to extract the handwritten numbers, the table data, and the logo image. Using Azure Content Understanding in Foundry Tools, what should they configure?

   • AA custom extraction pipeline with OCR for handwritten text, table extraction, and image extraction.
   • BA document classification pipeline that categorizes the sheet and extracts the table.
   • CAn attribute detection pipeline that extracts the handwritten numbers and the table.
   • DA summarization pipeline that summarizes the readings and includes the table data.
   Reveal Solution Discussion Select an answer first
6. 20application · medium

   A museum digitizes its archival documents, which include photographs, handwritten notes, and audio recordings of curators. They need to extract the photographs, detect the presence of a specific curator's signature, and process the audio recordings for archival. Using Azure Content Understanding in Foundry Tools, what should they build?

   • AA custom pipeline that accepts documents and audio, with image extraction and attribute detection for the signature.
   • BA summarization pipeline that summarizes the documents and includes the curator's signature.
   • CAn OCR pipeline to read the handwritten notes and then manually extract photos and signatures.
   • DA document classification pipeline that categorizes the archival items and extracts the photos.
   Reveal Solution Discussion Select an answer first